Research Methodology - Historical Land Use/Land Cover Classification Using Remote Sensing - page 97

5.5.3 The Choice of the most Appropriate Time to Obtain
Satellite Images/Temporal Considerations/Spatial
Aspects of Spectral Response Patterns

The questions related to temporal characteristics of the used data are: how much of
the remotely sensed data has to be obtained at a certain time? What time or times
of the year are more suitable (or required) for obtaining these remotely sensed
data? Are there particular daytime considerations that have to be involved during
the obtaining of data (Hoffer 1980 )? and, what are the spatial aspects of the
spectral response patterns?
Image acquisition date selection is essential for successful classification of
many vegetation covers, especially agricultural crops.
The study of LULC using remotely sensed data faces the problem of the
selection of the date in which the image was captured, i.e., the year and the month.
This selection is decisive with regard to the information which researchers receive.
Most of the irrigation projects discussed in this study were located within the five
agriculturally stable zones in Syria, which receive insufficient precipitation to
establish a rain-fed agriculture. Therefore, the majority of cultivated areas are
irrigated either in winter or in summer. The agricultural cycle of both winter and
summer crops ends in May and August, respectively. This means the spectral
differences reach their maximum point of clarity at this time, despite different
patterns of land use. For remote sensing based studies, the time of year of the
image capture is an important factor, because of the density of vegetation, both
natural and cultivated. This depends on many factors, notably the amount of
precipitation that changes from year to year; and human factors, such as the use of
fertilizers, which lead to changes in the characteristics of spectral reflectance/
response of a specific crop. For example, the spectral response of fertilized wheat
will differ from a field of the same crop which is unfertilized. The use of fertilizers
where insufficient water exists will lead to early yellowing of the crops.
World-wide, the best date range for identifying winter wheat is late March
through to early May, when the crop is at peak greenness. To identify corn and
other summer crops, the best date range is late July to mid-August. The most
important and best way to choose a proper time for remotely sensed data is to
study the growth stages of each type of vegetation and the spectral change in its
behavior during the months of growth, through field work and the use of spectral
reflectance measurement devices (Spectrometer) (see Sect. 5.5 ). As a result of
these measurements, the growth periods of a variety of crops and the differences in
their spectral reflections can be determined.
The results of the spectrometry readings taken for the purposes of this study are
outlined in Sect. 5.5 . When comparing the spectral reflectance curves of the
studied summer crops (cotton, and corn), it was found that the best spectral region
for the separation of crops was the near infrared domain. Under the conditions of
the project area, the best period to distinguish these differences was found to be the
period

between July 20 and August 20. Based on previous results, it was